---
title: InternScience/Sage-A1.5
url: https://recentmodel.com/models/InternScience/Sage-A1.5/
status: main board
published: 2026-10-15T08:34:39Z
updated: 2026-10-21T11:18:13Z
source: https://huggingface.co/InternScience/Sage-A1.5
---

# InternScience/Sage-A1.5

Sage‑A1.5 is a 35B Mixture‑of‑Experts agentic model based on Qwen3.6 and Agents-A1 from Shanghai AI Laboratory, built to advance long-horizon scientific problem solving through adaptive capability composition. Mixture-of-experts, 36B total (256 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.


- **Org:** InternScience
- **Released:** 2026-10-15
- **Status:** main board
- **License:** apache-2.0 (permissive)
- **Parameters:** 36.0B
- **Architecture:** MoE · 256 experts, 8/token · Qwen3_5MoeForConditionalGeneration
- **Context:** 262,144 tokens
- **Task:** text-generation · text
- **Signal:** 283 HF likes · 3875 downloads

## Runs on

real quant files; sizing calculation, not tested.

- Q4 ~23.4 GB: 32GB GPU (5090), Mac 36GB, Mac 48GB, CPU only, 32GB RAM (slow)
- Q8 ~39.5 GB: 48GB GPU (RTX 6000 Ada / A6000), Mac 64GB (M1 Max), Mac 96GB, CPU only, 64GB RAM (slow)
- FP16 ~72.8 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

## Links

- [Hugging Face](https://huggingface.co/InternScience/Sage-A1.5)
- [GGUF: InternScience/Sage-A1.5-F16-GGUF](https://huggingface.co/InternScience/Sage-A1.5-F16-GGUF)
- [quant: InternScience/Sage-A1.5-FP8](https://huggingface.co/InternScience/Sage-A1.5-FP8)
- [GGUF: InternScience/Sage-A1.5-Q4_K_M-GGUF](https://huggingface.co/InternScience/Sage-A1.5-Q4_K_M-GGUF)
- [GGUF: InternScience/Sage-A1.5-Q8_0-GGUF](https://huggingface.co/InternScience/Sage-A1.5-Q8_0-GGUF)
- [MLX: mlx-community/Sage-A1.5-OptiQ-4bit](https://huggingface.co/mlx-community/Sage-A1.5-OptiQ-4bit)